Google Project Skate

Project Skate

Google

The Google team wanted to capitalise on the growing interest of skateboarding ever since the 2021 Tokyo Olympic Games. Partnering with Nakatomi, they created an event-focused technical installation for half-pipe skateboarding, and collaborated with 14 year-old Olympic bronze medallist, Sky Brown. Using multiple cameras, Google's TensorFlow AI platform, and computer vision, the setup allowed both audience and skater to watch a data-augmented replay of their session, where tricks were automatically named and stats were generated.

Services

  • Machine Learning
  • Product Development
  • R&D

Problem to solve

Interest in skateboarding has soared, and as the sport's popularity skyrockets, so does its fanbase. Both skaters and spectators are eager to understand the sport more deeply.

Bold idea

Using machine learning to making skateboarding more accessible to the general public, by breaking down tricks into understandable components.

We’re excited to continue developing the technology and see how it can help showcase the performance of athletes, and bring equal participation across the board.

Tara McKenty
Google Creative Group Lead, Brand Studio APAC

How we got there

Skating in competitions since the age of eight, Sky Brown won a bronze medal at the 2021 Tokyo Olympics, representing Great Britain
Skating in competitions since the age of eight, Sky Brown won a bronze medal at the 2021 Tokyo Olympics, representing Great Britain

Closing the gap between the athletes and audience.

Skateboarding's recent inclusion as an Olympic sport has attracted talented newcomers from around the world to compete for gold. While the stunts these athletes perform are incredible, they're extremely challenging for beginners to replicate without achievable milestones.

The difficulty in explaining these performances is compounded by skateboarding tricks' impenetrable names, like the "Fakie McTwist" or "Frontside Caballerial." Often named after the first person to successfully perform them, these terms preserve the sport's history but remain unclear to casual observers. Even breaking down a trick's components involves jargon: frontside, backside, fakie, blunt, disaster, 180, and so on.

Three cameras were installed to ensure a view of the skateboarder from all angles.

Decoding the mechanics of skateboarding.

We used the open-source Google Tensorflow based AI system, BlazePose, along with studio-grade cameras and wireless sensors to track the pose of the skater and skateboard at sixty times per second.

We captured hundreds of example tricks on the Southern Hemisphere’s largest half-pipe with some of Australia’s best skateboarders at the Sydney Olympic Park. Hours of footage at 60fps from multiple angles let us see the athletes’ tricks in high detail. We hand labeled each of the tricks with useful information, like their approach side, the number of spins they performed, whether the trick was a grind or aerial, etc.

From analysing this footage, we realised that skateboarding tricks are composed of a small set of actions that are strung together in new and interesting ways. So, we developed a program to measure some of the fundamental parts of the skateboard trick, like: how fast the skateboarder was going, how high they reached, how many rotations they performed, where they grabbed the board, and how the board moved in the air or on the rail.

When we had all the separate parts, we combined these systems together and had the ability to name the tricks ourselves. To make the names more friendly and universally accepted, our skateboarders recommended that we take them from the Tony Hawk Pro Skater video game. Which we then played a lot of — for research, of course.

People don’t really look at height right now. They just look at difficulty. This is going to push me.

Sky Brown
Pro Skater

Building and testing the system with Sky Brown.

We developed a working prototype on-set in San Francisco just nine weeks after receiving the initial brief. Sky Brown, the 14-year-old Olympic bronze medalist in women's street skateboarding, became our beta tester.

Google collaborated with Vice to produce a film showcasing this technology. The film was presented at the Google Developers Conference, and received significant interest from both internal and external business stakeholders.

Project Skate enables skateboarders – who may be underestimated due to human bias – to demonstrate the extent of their skills with certainty.

David Arcus
Google Creative Technologist, Brand Studio APAC
Corporates
All Work

Have an uncommon problem or promising pitch?
Let’s partner up.

Get in touch